跳到正文
arXiv cs.AI· Zhihua Wang, Yifeng Gao, Yunhao Feng, Rui Pan, Zhouyuan Ma, Baiheng Zhao, Zixing Chen, Yanping Li, Yizhang Liu, Xingjun Ma·· 2 小时前AI 评分35

SFMambaNet:面向双视图对应学习的频率感知 Mamba

Smooth Inliers, Erratic Outliers: Frequency-Aware Mamba for Two-View Correspondence Learning

AI 导读

研究者提出 SFMambaNet,按簇扫描候选匹配并用前向与反向传递之间的可学习频域门控过滤特征,用于双视图对应学习(对应点剔除)。该方法在相机位姿估计、外点剔除和单应性估计上优于约 200 万参数规模的空间域 Mamba 基线。实验还发现无参数低通滤波器优于同等规模的逐点层,学习到的门控会自发转为低通,使扫描对匹配顺序打乱不敏感。

正文 · 原文

View PDF HTML (experimental)

Abstract:When shown matches between two photographs of the same cathedral, people can tell at a glance which ones are wrong: a window matched to its look-alike jumps against its neighbors, while correct matches shift together. Two-view correspondence learning, also called correspondence pruning, asks a network to make this judgment for thousands of candidate matches, most of which are wrong. Mamba, a fast sequence model, now drives strong pruners, but it reads the matches as a plain sequence, and whether its scan can capture such agreement is unclear. We propose SFMambaNet, which scans the matches cluster by cluster, so that correct matches vary slowly along the sequence while wrong ones stay erratic, and filters the scanned features with learnable frequency-domain gates between its forward and backward passes. On camera pose estimation, outlier removal and homography estimation, we find that SFMambaNet outperforms spatial-domain Mamba baselines with about two million parameters. Contrary to the usual expectation that such gains come from extra capacity, we find that a parameter-free low-pass filter beats point-wise layers as large as the gates. The learned gates turn low-pass on their own and make the scan far less sensitive to how the matches are shuffled. Together, these results show that frequency is a natural language for describing agreement between correspondences, with implications for other sequence models that must read unordered sets, such as point clouds and graphs.
Comments: 26 pages, including appendices. Substantially revised and expanded version with updated title, author list, analysis, experiments, and supplementary material
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as: arXiv:2606.04493 [cs.CV]
  (or arXiv:2606.04493v2 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2606.04493

arXiv-issued DOI via DataCite

Submission history

From: Zhihua Wang [view email]
[v1] Wed, 3 Jun 2026 06:18:05 UTC (40,303 KB)
[v2] Tue, 17 Nov 2026 17:32:33 UTC (5,662 KB)

来源:arXiv cs.AI · arxiv.org